sparcl (version 1.0.3)

ColorDendrogram: Color the leaves in a hierarchical clustering dendrogram

Description

Pass in the output of "hclust" and a class label for each observation. A colored dendrogram will result, with the leaf colors indicating the classes.

Usage

ColorDendrogram(hc, y, main = "", branchlength = 0.7, labels = NULL, xlab = NULL, sub = NULL, ylab = "", cex.main = NULL)

Arguments

hc

The output of running "hclust" on a nxn dissimilarity matrix

y

A vector of n class labels for the observations that were clustered using "hclust". If labels are numeric from 1 to K, then colors will be determine automatically. Otherwise the labels can take the form of colors (e.g. c("red", "red", "orange", "orange")).

main

The main title for the dendrogram.

branchlength

How long to make the colored part of the branches. Adjustment will be needed for each dissimilarity matrix

labels

The labels for the n observations.

xlab

X-axis label.

sub

Sub-x-axis label.

ylab

Y-axis label.

cex.main

The amount by which to enlarge the main title for the figure.

References

Witten and Tibshirani (2009) A framework for feature selection in clustering.

See Also

HierarchicalSparseCluster, HierarchicalSparseCluster.permute

Examples

# Generate 2-class data
set.seed(1)
x <- matrix(rnorm(100*20),ncol=20)
y <- c(rep(1,50),rep(2,50))
x[y==1,] <- x[y==1,]+2
# Perform hierarchical clustering
hc <- hclust(dist(x),method="complete")
# Plot
ColorDendrogram(hc,y=y,main="My Simulated Data",branchlength=3)